Grand mal

European Medicines Board was been having a field-day or open-house in London yesterday taking depositions and gathering evidence about Valproate aka 2-propyl-valeric acid which has been used in medicine since about 1962. There's 5 hours of it captured on youtube.  It's a simple compound: a carboxylic acid with two tails like a bigger fancier acetic acid.  It is marketed by French megapharm Sanofi as the sodium salt, often under the brand name Epilim and because it is long off patent, the daily dose regime is affordable in the Western World - maybe $1 a day but 80% cheaper in the Third World. Where? When? To whom? Primarily to epileptics where it has been found very effective in preventing seizures. But doctors have been prescribing it for migraines and bipolar disorder . . . because they can. Nobody really knows how valproate works - maybe something to do with affecting GABA the neurotransmitter - but it is known to have a number of side-effects. You would put up with a bit of nausea, drowsiness, and dry-mouth to ward off an epileptic seizure which is no fun at all. Even adverse effects on liver-function would probably be tolerable in the utilitarian balance.  But if the cost of avoiding a seizure through valproate use was a much higher likelihood of delivering a child with neurological defect like spina bifida, or one cognitively impaired, or 10x more likely to be autistic; then maybe you'd prefer the risk of seizures for the few months long long months of pregnancy. a

The under-appreciated problem of the pharmaceutical world is the interaction terms. The drug company may have carried out all the efficacy and side-effects tests in healthy male volunteers but it would be absurd to require trials of valproate in combination with all 10,000 different drugs in the pharmacopoeia: nothing would ever get to market if you so insisted. A whole alphabet of other active compounds have been shown to synergistic effects with valproate: Aspirin: Benzodiazepines: Carbapenems: Cimetidine: Erythromycin: Ethosuximide: Felbamate: Mefloquine: Primidone: Rifampin: Warfarin: Zidovudine. The likelihood of an adverse effect on the foetus is increased a) if the Epilim dose is higher but also b) if other drugs have been taken 'in cocktail'.

We do these cost-benefit reckonings all the time if we are healthy. My drug of choice is aspirin, I take about 2 a year to drive off a headache, which is thankfully that rare for me. I've got a job, so the cost of the aspirin is nothing-at-all compared to the relief it brings. If it cost €500 a pop, though, I would probably tough out the headache and trust it would be gone by morning.  If we are sick (and therefore vulnerable, distracted and anxious), we rely on our doctor to prescribe medicines at appropriate (hopefully minimal-but-effective dose) appropriate to our age, sex and condition. Many of us run a cross-check with Dr Google, but the long words on the internet bamboozle us more effectively than a doctor trying to get through his list and out to the golf-course. It is, for example, a bit disconcerting to be reading the small print of one of Pat the Salt's medications to find that it is not recommended for the elderly because it greatly increases the risk of stroke.

Within ten years of launch in the pharmacopoeia, epidemiologists were noticing that epileptic women were presenting at doctors surgeries with neurologically compromised children. But Epilim was so effective for epilepsy that these data were suppressed from general release and only doctors were warned about the possibility of adverse effect on the foetus. I can only imagine the amount to paperwork that GPs are expected to read, learn and inwardly digest so that they can remember which drugs have which side-effects with which classes of patient. Bearing in mind that every hour of surgery 10 or even 20 people will appear with a bewildering array of symptoms. Nevertheless for 40 years, these things have been known, and in effect suppressed. Check out the Irish HSE fact-sheet on this drug.

Last year, a study in Rhone-Alpes found a connexion between the drug, marketed as Depakine, and the birth defects. Extrapolating to metropolitan France indicated that as many as 450 such events had occurred between 2006 and 2014. The authorities mandated more obvious labelling on the packet. In Ireland, record-checking in Ireland suggested that 1,600 women of child-bearing age were prescribed valproate over a similar period. But writing warnings requires that the user is not ditracted by pain and anxiety and has good enough eyesight to read the small print. Every adverse side-effect is typically recorded without clearly indicating which ones have the greatest risk = magnitude of badness x likelihood of occurrence. Which is fair enough because your risk is not my risk - I am not, nor have I ever been, pregnant, for example.

Sanofi has been selling this stuff for 40 years and seem to have altered the labelling only when required to do so by a government enquiry - changing the packaging and reprinting the Advice adversely affects the bottom line. A class-action suit in 2010 fizzled out because the legal costs couldn't be guaranteed to the prosecuting lawyers. This new publicity and documentation will have lawyers all over the World cuing up offering their services no-win, no-fee, because the treasure ship SS Sanofi is now ready for taking . . . and thousands of mothers will get some cash to help them cope with their many troubled children.

I should add that this story only made the Irish news because an Irish mother of three affected boys took up a chunk of the 5 hours of testimony at the EMA hearing. The news is always parochial.